I am trying to develop a sequential CNN model for multi-class image classification. After developing the model, when I started to train, the model gave accuracy in the very first epochs, but 2nd epochs i was not getting any accuracy. But in 3rd epoch, I am getting accuracy again. Why this is happening? One Additional Question, Can you suggest to me how can i improve my f1 score, precision, recall of the model.[enter image description here](https://i.sstatic.net/Jpn1fW72.png)
I have tryed ChatGPT to solve this problem , but it cant find any solution.